|
|
WEAT: Web Enabled Analysis Tool
The BRFSS, the world’s largest telephone survey, tracks health risks in the United States. Information from the survey is used to improve the health of the American people. |
Logistic Regression Information
EXAMPLE:
Logistic regression allows you to generate models that predict a binary dependent variable from one or more independent, or predictor, variables. For example, imagine that you are interested in learning the impact of tobacco use and exercise on obesity risk. The odds ratios produced in a logistic regression model will help you interpret the direction and degree of influence of these predictors.
To generate a model to help answer this question, you would follow these steps:
- Go to the Logistic Regression page and under "Step 1 Select Dependent Variable" select "Risk factor: Overweight or Obese."
- Under "Step 2 Select Independent Variables" select "Exercise."
- A new phrase appears on-screen, "Exercise: Exercised in past 30 days" beside a checkbox.
- Click in the checkbox to save that independent variable to the frame "Equation Viewer."
- Under "Step 2 Select Independent variables" select "Tobacco."
- A new phrase appears on-screen, "Tobacco: currently smoking."
- Check the checkbox to save that independent variable to the frame "Equation Viewer."
- Click on "Run Report."
- Review the resulting regression model on the screen. Note that the odds ratio for obesity among those who exercised in the past 30 days is below 1.00, meaning that their risk is less than those who have not exercised.
Every time you want to return to the Logistic Regression page to add or change variables, click on the button "Modify the Logistic Model," or click on the back arrow on your browser. Your original selected variables will be available for you to modify as needed. If you want to download results to a spreadsheet, click on the button "Download to Excel Spreadsheet."
The WEAT allows users to change the reference groups for some independent variables. At the bottom of the screen, simply click the check box beside "Step 3 Reference Categories." A new screen will appear that reads "Step 4 Reference Categories Selection." From there you can select the desired level of any categorical variables for your reference variable and click on "Run Report" to see the results.
Back to top
|
|